KARATECH: A Practice Support System Using an Accelerometer to Reduce the Preliminary Actions of Karate.

Kumite is a karate sparring competition in which two players face off and perform offensive and defensive techniques. Depending on the players, there may be preliminary actions (hereinafter referred to as "pre-actions"), such as pulling the arms or legs, lowering the shoulders, etc., just before a technique is performed. Since the presence of a pre-action allows the opponent to know the timing of the technique, it is important to reduce pre-actions in order to improve the kumite. However, it is difficult for beginners and intermediate players to accurately identify their pre-actions and to improve them through practice. Therefore, this study aims to construct a practice support system that enables beginners and intermediate players to understand their pre-actions. In this paper, we focus on the forefist punch, one of kumite's punching techniques. We propose a method to estimate the presence or absence of a pre-action based on the similarity between the acceleration data of an arbitrary forefist punch and a previously prepared dataset consisting of acceleration data of the forefist punch without a pre-action. We found that the proposed method can estimate the presence or absence of a pre-action in an arbitrary forefist punch with an accuracy of 86%. We also developed KARATECH as a system to support the practice of reducing pre-actions using the proposed method. KARATECH shows the presence or absence of pre-actions through videos and graphs. The evaluation results confirmed that the group using KARATECH had a lower pre-action rate.

Physical Noninvasive Attacks on Photoplethysmogram by Computer Controlled Blood Pressure Cuff.

Sensor data has been used in social security and welfare infrastructures such as insurance and medical care to provide personalized products and services; there is a risk that attackers can alter sensor data to obtain unfair benefits. We consider that one of the attack methods to modify sensor data is to attack the wearer's body to modify biometric information. In this study, we propose a noninvasive attack method to modify the sensor value of a photoplethysmogram. The proposed method can disappear pulse wave peaks by pressurizing the upper arm with air pressure to control blood volume. Seven subjects experiencing a rest environment and five subjects experiencing an after-exercise environment wore five different models of smartwatches, and three pressure patterns were performed. It was confirmed in both situations that the displayed heart rate decreased from the true heart rate.

Estimating Scalp Moisture in a Hat Using Wearable Sensors.

Hair quality is easily affected by the scalp moisture content, and hair loss and dandruff will occur when the scalp surface becomes dry. Therefore, it is essential to monitor scalp moisture content constantly. In this study, we developed a hat-shaped device equipped with wearable sensors that can continuously collect scalp data in daily life for estimating scalp moisture with machine learning. We established four machine learning models, two based on learning with non-time-series data and two based on learning with time-series data collected by the hat-shaped device. Learning data were obtained in a specially designed space with a controlled environmental temperature and humidity. The inter-subject evaluation showed a Mean Absolute Error (MAE) of 8.50 using Support Vector Machine (SVM) with 5-fold cross-validation with 15 subjects. Moreover, the intra-subject evaluation showed an average MAE of 3.29 in all subjects using Random Forest (RF). The achievement of this study is using a hat-shaped device with cheap wearable sensors attached to estimate scalp moisture content, which avoids the purchase of a high-priced moisture meter or a professional scalp analyzer for individuals.

System for Detecting Learner Stuck in Programming Learning.

Getting stuck is an inevitable part of learning programming. Long-term stuck decreases the learner's motivation and learning efficiency. The current approach to supporting learning in lectures involves teachers finding students who are getting stuck, reviewing their source code, and solving the problems. However, it is difficult for teachers to grasp every learner's stuck situation and to distinguish stuck or deep thinking only by their source code. Teachers should advise learners only when there is no progress and they are psychologically stuck. This paper proposes a method for detecting when learners get stuck during programming by using multi-modal data, considering both their source code and psychological state measured by a heart rate sensor. The evaluation results of the proposed method show that it can detect more stuck situations than the method that uses only a single indicator. Furthermore, we implemented a system that aggregates the stuck situation detected by the proposed method and presents them to a teacher. In evaluations during the actual programming lecture, participants rated the notification timing of application as suitable and commented that the application was useful. The questionnaire survey showed that the application can detect situations where learners cannot find solutions to exercise problems or express them in programming.

Investigation of the Effect of Students' Nodding on Their Arousal Levels in On-Demand Lectures.

Due to COVID-19, various lecture styles are being explored. On-demand lectures are attracting increasing attention due to advantages such as being able to watch without restrictions due to location and time. In contrast, on-demand lectures have disadvantages, such as no interaction with the lecturer, so the quality of on-demand lectures should be improved. Our previous study showed that when participants nod without showing their faces in a real-time remote lecture, their heart rate state changes to arousal and nodding can increase arousal. In this paper, we hypothesize that nodding during on-demand lectures increases participants' arousal levels, and we investigate the relationship between natural and forced nodding and the level of arousal based on heart rate information. Students taking on-demand lectures rarely nod naturally, so we used entrainment to encourage nodding by showing a video of another participant nodding, and by forcing the participants to nod when the other participant nodded in the video. The results showed that only participants who nodded spontaneously changed the value of pNN50, an index of the arousal level, which reflected a state of high arousal after one minute. Thus, participants' nodding in on-demand lectures can increase their arousal levels; however, the nodding must be spontaneous, not forced.

Method for Recognizing Pressing Position and Shear Force Using Active Acoustic Sensing on Gel Plates.

A touch interface is an important technology used in many devices, including touch panels in smartphones. Many touch panels only detect the contact position. If devices can detect shear force in addition to the contact position, various touch interactions are possible. We propose a two-step recognition method for recognizing the pressing position and shear force using active acoustic sensing, which transmits acoustic signals to an object and recognizes the state of the object by analyzing its response. Specifically, we attach a contact speaker transmitting an ultrasonic sweep signal and a contact microphone receiving ultrasonic waves to a plate of gel. The propagation characteristics of ultrasonic waves differ due to changes in the shape of the gel caused by the user's actions on the gel. This system recognizes the pressing position and shear force on the basis of the difference in acoustic characteristics. An evaluation of our method involving a user-independent model confirmed that four pressing positions were recognized with an F1 score of 85.4%, and four shear-force directions were recognized with an F1 score of 69.4%.

Unconscious Other's Impression Changer: A Method to Manipulate Cognitive Biases That Subtly Change Others' Impressions Positively/Negatively by Making AI Bias in Emotion Estimation AI.

Artificial Intelligence (AI) for human emotion estimation, such as facial emotion estimation, has been actively studied. On the other hand, there has been little research on unconscious phenomena in cognition and psychology (i.e., cognitive biases) caused by viewing AI emotion estimation information. Therefore, this study verifies RQ "Do people have a cognitive bias in which impressions of others (i.e., how to see and feel about others) are changed by viewing biased AI's emotion estimation information? If it exists, can impression manipulation methods that intentionally use this cognitive bias be realized?" The proposed method for verification makes the emotion estimation system biased so as to estimate emotion more positively/negatively than AI without bias. A prototype system was implemented. Evaluation using video showed that the presentation of biased emotion estimation information causes a phenomenon that quickly and unconsciously changes the way people see and feel others' impressions, which supported the RQ. Specifically, viewing information that estimated others' emotions more positively/negatively caused the phenomenon in which the user's self-judgment was overridden and others' impressions of emotions, words, and actions were perceived more positively/negatively. The existence of this phenomenon and method indicates that biased emotion estimation AI has the potential to both cause adverse effects on people and support people for good purposes through the manipulation of their impressions. This study provides helpful insights for the design and use of emotion estimation AI considering cognitive biases.

Extensible Neck: A Gesture Input Method to Extend/Contract Neck Virtually in Video See-through AR Environment.

With the popularization of head-mounted displays (HMDs), many systems for human augmentation have been developed. This will increase the opportunities to use such systems in daily life. Therefore, the user interfaces for these systems must be designed to be intuitive and highly responsive. This paper proposes an intuitive input method that uses natural gestures as input cues for systems for human augmentation. We investigated the appropriate gestures for a system that expands the movements of the user's viewpoint by extending and contracting the neck in a video see-through AR environment. We conducted an experiment to investigate natural gestures by observing the motions when a person wants to extend his/her neck. Furthermore, we determined the operation method for extending/contracting the neck and holding the position through additional experiments. Based on this investigation, we implemented a prototype of the proposed system in a VR environment. Note that we employed a VR environment since we could test our method in various situations, although our target environment is AR. We compared the operability of the proposed method and the handheld controller using our prototype. The results confirmed that the participants felt more immersed using our method, although the positioning speed using controller input was faster than that of our method.

The structure of the superior and inferior parietal lobes predicts inter-individual suitability for virtual reality.

The global virtual reality (VR) market is significantly expanding and being challenged with an increased demand owing to COVID-19. Unfortunately, VR is not useful for everyone due to large interindividual variability existing in VR suitability. To understand the neurobiological basis of this variability, we obtained neural structural and functional data from the participants using 3T magnetic resonance imaging. The participants completed one of two tasks (sports training or cognitive task) using VR, which differed in the time scale (months/minutes) and domain (motor learning/attention task). Behavioral results showed that some participants improved their motor skills in the real world after 1-month training in the virtual space or obtained high scores in the 3D attention task (high suitability for VR), whereas others did not (low suitability for VR). Brain structure analysis revealed that the structural properties of the superior and inferior parietal lobes contain information that can predict an individual's suitability for VR.

A Method for Behavior Change Support by Controlling Psychological Effects on Walking Motivation Caused by Step Count Log Competition System.

Systems presenting information that encourages competition by using rankings and scores (hereafter referred to as competition information) have become widespread to support behavioral change. However, users without high levels of motivation, such as behavior change support targets, do not necessarily benefit from competition information. In this study, we propose a method to control the psychological effects caused by competition information to support behavior change. We implemented a competition information presentation system using step counts logs to support increasing one's daily steps. We designed two patterns of competition information considering psychological effects. One is likely to have good effects, using three mechanisms to easily obtain results corresponding to the effort, make closely matched rivals with similar abilities, and pay attention to a small number of rivals. The other is unlikely to have positive effects and may potentially even have negative ones, using a mechanism that brings about the opposite results of the former pattern. We evaluated 42 participants with low levels of motivation over six weeks. The results showed that the former information pattern increased participants' step counts by about 1000 steps per day, and the latter information pattern did not lead to an increase. We confirmed the feasibility of the proposed method and discussed the possibility of the appropriate use and potential abuse of such techniques for manipulating motivation. Our study can be helpful in designing a competition information presentation system considering psychological effects.

Effectiveness of time-varying echo information for target geometry identification in bat-inspired human echolocation.

Bats use echolocation through flexible active sensing via ultrasounds to identify environments suitable for their habitat and foraging. Mimicking the sensing strategies of bats for echolocation, this study examined how humans acquire new acoustic-sensing abilities, and proposes effective strategies for humans. A target geometry identification experiment-involving 15 sighted people without experience of echolocation-was conducted using two targets with different geometries, based on a new sensing system. Broadband frequency-modulated pulses with short inter-pulse intervals (16 ms) were used as a synthetic echolocation signal. Such pulses mimic buzz signals emitted by bats for echolocation prior to capturing their prey. The study participants emitted the signal from a loudspeaker by tapping on Android devices. Because the signal included high-frequency signals up to 41 kHz, the emitted signal and echoes from a stationary or rotating target were recorded using a 1/7-scaled miniature dummy head. Binaural sounds, whose pitch was down-converted, were presented through headphones. This way, time-varying echo information was made available as an acoustic cue for target geometry identification under a rotating condition, as opposed to a stationary one. In both trials, with (i.e., training trials) and without (i.e., test trials) answer feedback immediately after the participants answered, the participants identified the geometries under the rotating condition. Majority of the participants reported using time-varying patterns in terms of echo intensity, timbre, and/or pitch under the rotating condition. The results suggest that using time-varying patterns in echo intensity, timbre, and/or pitch enables humans to identify target geometries. However, performance significantly differed by condition (i.e., stationary vs. rotating) only in the test trials. This difference suggests that time-varying echo information is effective for identifying target geometry through human echolocation especially when echolocators are unable to obtain answer feedback during sensing.

Monitoring respiratory rates with a wearable system using a stretchable strain sensor during moderate exercise.

Respiratory rate, a sensitive indicator of respiratory status, is rarely measured during the field walking test. Our objective was to develop and validate a non-invasive, wearable monitoring system using stretchable strain sensors and an accompanying algorithm capable of providing real-time measurements of respiration during exercise. Twenty-four healthy volunteers wore stretchable sensors during a walking test protocol that included standing, sitting, walking, and walking with a stick. Sensors were placed on the ribcage and abdomen. The Bland-Altman method was used to assess the accuracy and precision of breath counts; total respiration time and inspiration time ratio were determined by custom algorithms and compared with measurements obtained with the standard flow sensor. The output signal from the stretchable sensor was highly synchronized with flow signals. The limits of agreement were within 3 breaths/min throughout the test protocol. Differences between sensors for total respiration time and inspiration time ratio were less than 14% and 26%, respectively. The agreement was maintained regardless of respiratory rate or volume. The wearable respiratory monitoring system yielded accurate and precise breath counts and total duration of respiratory cycle during moderate exercise in healthy young individuals, suggesting that it might be useful in clinical practice. Graphical abstract.

Evaluation on Context Recognition Using Temperature Sensors in the Nostrils.

We can benefit from various services with context recognition using wearable sensors. In this study, we focus on the contexts acquired from sensor data in the nostrils. Nostrils can provide various contexts on breathing, nasal congestion, and higher level contexts including psychological and health states. In this paper, we propose a context recognition method using the information in the nostril. We develop a system to acquire the temperature in the nostrils using small temperature sensors connected to glasses. As a result of the evaluations, the proposed system can detect breathing correctly, workload at an accuracy of 96.4%, six behaviors at an accuracy of 54%, and eight behaviors in daily life at an accuracy of 86%. Moreover, the proposed system can detect nasal congestion, therefore, it can log nasal cycles that are considered to have a relationship with the autonomic nerves and/or biological states.